In the present study we provide evidence that SRP-35, a protein we identified in rabbit skeletal muscle sarcoplasmic reticulum, is an all-trans-retinol dehydrogenase. Analysis of the primary structure and tryptic digestion revealed that its N-terminus encompasses a short hydrophobic sequence bound to the sarcoplasmic reticulum membrane, whereas its C-terminal catalytic domain faces the myoplasm. SRP-35 is also expressed in liver and adipocytes, where it appears in the post-microsomal supernatant; however, in skeletal muscle, SRP-35 is enriched in the longitudinal sarcoplasmic reticulum. Sequence comparison predicts that SRP-35 is a short-chain dehydrogenase/reductase belonging to the DHRS7C [dehydrogenase/reductase (short-chain dehydrogenase/reductase family) member 7C] subfamily. Retinol is the substrate of SRP-35, since its transient overexpression leads to an increased production of all-trans-retinaldehyde. Transfection of C2C12 myotubes with a fusion protein encoding SRP-35–EYFP (enhanced yellow fluorescent protein) causes a decrease of the maximal Ca2+ released via RyR (ryanodine receptor) activation induced by KCl or 4-chloro-m-chresol. The latter result could be mimicked by the addition of retinoic acid to the C2C12 cell tissue culture medium, a treatment which caused a significant reduction of RyR1 expression. We propose that in skeletal muscle SRP-35 is involved in the generation of all-trans-retinaldehyde and may play an important role in the generation of intracellular signals linking Ca2+ release (i.e. muscle activity) to metabolism.

INTRODUCTION

The skeletal muscle SR (sarcoplasmic reticulum) is an intracellular membrane compartment highly specialized in calcium homoeostasis. Structurally it can be subdivided into two membrane portions: the terminal cisternae facing the transverse tubules and the LSR (longitudinal SR) that connects two terminal cisternae [1]. Depolarization of the plasma membrane causes the Ca2+ stored in the SR to be released, leading to muscle contraction by a process known as excitation–contraction coupling [2,3]; Ca2+ is then pumped back into the SR via SERCAs (sarcoplasmic/endoplasmic reticulum Ca2+-ATPases), which are located on the terminal cisternae and LSR, leading to muscle relaxation [46]. These highly co-ordinated events occur within milliseconds of each other thanks to the spatial organization of the membrane compartments carrying the different proteins involved in sensing plasma membrane depolarization, Ca2+ release and Ca2+ uptake [7]. Indeed, excitation–contraction coupling occurs at the triad, a structure made up of the transverse tubules (which are invaginations of the plasma membrane), carrying the voltage-sensing DHPR (dihydropyridine receptor) and two terminal cisternae carrying the RyR (ryanodine receptor) Ca2+ release channel [8,9]. Although these two Ca2+ channels are the basic unit underlying excitation–contraction coupling, they nevertheless function in co-ordination with a number of accessory proteins and enzymes involved in maintaining the architecture and optimal function of the calcium release unit. Because of their potential roles in regulating excitation–contraction coupling and since they may be targets of mutations causing neuromuscular disorders, several laboratories have focused their research on identifying novel proteins present on these specialized intracellular membrane compartments and a number of minor components, including mitsugumins, junctophilins, junctate, JP-45 and SRP27, have been characterized (for reviews, see [10,11]).

In the present study, we identified a 35 kDa protein using a proteomic approach in skeletal muscle SR, which we called SRP-35, for SR protein of 35 kDa. Sequence motif comparison indicates that this protein belongs to the DHRS7C {dehydrogenase/reductase [SDR (short-chain dehydrogenase/reductase) family] member 7C} subfamily [12] and is likely to catalyse the conversion of all-trans-retinol to retinaldehyde by reducing the cofactor NAD+ to NADH [13]. SDRs are ubiquitously expressed and constitute a large protein family involved in the reduction of a variety of substrates, among which is retinol, the precursor of retinoic acid [14]. In vertebrates, the effectors of retinoic acid signalling belong to the nuclear receptor family that are divided into two subgroups: the nuclear RARs (retinoic acid receptors), RARα/β/γ, and the RXRs (retinoid X receptors), RXRα/β/γ [1518]. Upon retinoic acid binding to RAR, the receptors undergo hetero-dimerization with RXRs and translocate into the nucleus, where they bind to target DNA sequences known as retinoic-acid-response elements, promoting the transcription of several genes, including some involved in growth, development, differentiation, cytokine production and metabolism [14,19]. Interestingly, treatment of mice with all-trans-retinoic acid reduces body weight, adipose tissue content and promotes skeletal muscle fatty acid oxidation [20]. Thus SRP-35 may be part of a signalling pathway linking muscle activity and metabolism; indeed, this polypeptide is enriched in tissues involved in fatty acid metabolism, notably liver, adipose tissue, kidney and skeletal muscle. Interestingly, in the latter tissue it is enriched in the LSR, where a number of glycolytic enzymes, such as pyruvate kinase, aldolase, enolase and glyceraldehyde 3-phosphate dehydrogenase have also been found to compartmentalize [21]. Since SRP-35 oxidizes all-trans-retinol to all-trans-retinaldehyde at a site directly involved in Ca2+ homoeostasis, we hypothesize that SRP-35 might have a dual function: first, reducing NAD+ linked to lactic acid generation. The NADH would in turn down-regulate Ca2+ release by affecting the redox equilibrium adjacent to RyR1 (the skeletal muscle isoform of the RyR). Secondly, SRP may regulate skeletal muscle gene transcription via the RAR pathway.

EXPERIMENTAL

Materials

Anti-calregulin and anti-calnexin antibodies, peroxidase-conjugated protein A, trypsin-EDTA, DMEM (Dulbecco's modified Eagle's medium), MEM (minimal essential medium) plus Earls salts, L-glutamine, penicillin/streptomycin, horse serum, fetal bovine serum, deoxyribonuclease I, Alexa Fluor® 488 chicken anti-rabbit IgG and Alexa Fluor® 568 donkey anti-mouse IgG were from Invitrogen. All-trans-retinol and all-trans-retinaldehyde, Protein A–Sepharose, ESCORT IV, peroxidase-conjugated Protein G, peroxidase-conjugated anti-mouse IgG and rabbit anti-DHRS7C anitbodies were from Sigma–Aldrich. Mouse anti-RyR1 (MA3-925) and anti-SERCA (MA3-910) monoclonal antibodies were from Thermo Scientific. The pGEX-5X-3 plasmid and HiTrap Blue Sepharose were from GE Healthcare. The UNI-ZAP XR cDNA library, pEYFP-C1 [where EYFP is enhanced YFP (yellow fluorescent protein)] and pEGFP-C1 [where EGFP is enhanced GFP (green fluorescent protein)] plasmids were from Clontech. Goat anti-DHPRα1.1 antibody was from Santa Cruz Biotechnology. Anti-albumin antibodies were from Bethyl Laboratories. Oxy-Blot protein oxidation detection kit was from Millipore. Nitrocellulose membrane was from Schleicher & Schuell BioScience. Protein molecular mass markers and the Protein Assay Kit were from Bio-Rad Laboratories. NAD+, NADH, NADP+, FuGENE™ 6, anti-proteases and Taq polymerase were from Roche Diagnostics. The High Capacity cDNA RT (reverse transcription) kit and fast SYBR Green master mix were from Applied Biosystems. Fura 2 acetoxymethylester was from Calbiochem. Tri-Reagent was from Molecular Research Center. PCR primers were from Microsynth. Paraffin-embedded mouse skeletal muscle sections were from Ciochain Institute (BioCat). All other chemicals were reagents of the highest available grade. Mice were bred in-house and rabbits (New Zealand White) were purchased from Charles Rivers Laboratories. All procedures were performed in accordance with the stipulations of the Helsinki Declarations for care and use of laboratory animals. The experiments were approved by the local Basel Kantonal authorities to use animals as tissue donors and for antibody production.

Protein sequencing

Proteins present in the terminal cisternae and junctional face membrane fractions were separated on SDS/PAGE (10% gels) prepared under ultraclean conditions and allowed to polymerize at room temperature (23°C) for 72 h; the gel was stained with Coomassie Brilliant Blue R250 [0.1% in 50% (v/v) methanol and 0.1% acetic acid] and destained overnight at 4°C in 1% acetic acid and 50% (v/v) methanol; the 35 kDa band was cut out with a clean scalpel blade and the SDS and Coomassie Brilliant Blue were removed by slicing the gel finely in 40% (v/v) propan-1-ol, washing and vortexing the sample and changing the solution several times. After the last propan-1-ol wash, the gel was resuspended in a solution containing 0.2 M NH4HCO3 and 40% (v/v) acetonitrile and washed with this solution 5 times. After the final wash, excess liquid was removed and the gel fragments were dried in a speedvac. The sample was then rehydrated in 0.1 M NH4CO3 and trypsin (sequencing grade, 25 ng/μl) and incubated at 37°C overnight. The supernatant was then collected and transferred to a clean tube; the pellet was treated with a solution made with 80% (v/v) acetonitrile and 0.1% trifluoroacetic acid to extract the remaining peptides. The two supernantants were combined and dried in a speedvac and then subjected to MS analysis using a Q-TOF (quadrupole–time-of-flight; Micromass) mass spectrometer [22].

SRP-35 cDNA cloning

Total RNA was isolated from mouse skeletal muscles using Tri-Reagent following the instructions provided by the manufacturer (Molecular Research Center). Total RNA was converted to cDNA as described previously [23] and primers were designed based on the peptide sequence obtained from trypsin digestion of rabbit SRP-35. These primers (5′-GGAGAGCCTCTACGCTGCCT-3′ and 5′-ACGCCAACTACTTTGGACCCA-3′ for forward and reverse reactions respectively) were used to amplify a sequence from mouse skeletal muscle cDNA; a band of approximately 220 bp was obtained using the following amplification conditions: 1 cycle at 95°C for 5 min followed by 35 cycles of annealing (55°C for 30 s), extension (72°C for 45 s) and denaturation (95°C for 45 s), followed by a 5 min extension cycle at 72°C. The PCR-amplified cDNA was then used as a probe to screen a mouse skeletal muscle UNI-ZAP XR cDNA library as described previously [23]. A clone of 960 bp, containing the entire ORF (open reading frame), was pulled out and sequenced (Microsynth). The coding sequence of SRP-35 was subcloned into the pGEX-5X-3 bacterial expression plasmid and into pEGFPC1 and pEYFPC1 for expression in mammalian cells.

Cell culture and transfection

C2C12 cells were cultured in growth medium [DMEM plus glutamax, high glucose (4.5 g/l), 20% (v/v) fetal bovine serum, 200 mM L-glutamine and penicillin/streptomycin (50 units/ml and 50 μg/ml final concentration respectively)] and maintained below confluence. For differentiation into myotubes, cells were allowed to reach 70% confluence, then the medium was switched to the differentiation medium [DMEM plus Glutamax and high glucose, 5% (v/v) horse serum, 200 mM L-glutamine and penicillin/streptomycin), until cells had visibly fused into myotubes (approximately 5 days). Transfection of C2C12 cells was carried out using FuGENE™ 6 at a ratio of 1.5 μl of FuGENE™ per μg of plasmid DNA per ml culture media. Cells were transfected on days 1 and 3 following the switch to differentiation medium and harvested on day 4 post-differentiation. In some experiments, C2C12 were incubated for 4 days with retinoic acid (5 μM) during differentiation.

HEK-293 (human embryonic kidney 293) cells were cultured and transfected with the pEGFPC1 and pEGFPC1-SRP-35-plasmids using ESCORT IV as described previously [24].

Subcellular fractionation, trypsin digestion and fluorescence analysis

Total microsomes from different mouse tissues (heart, lung, brain, kidney, skeletal muscle, liver, spleen, stomach and intestine), total homogenates prepared from mouse tissues and from isolated fast [EDL (extensor digitorum longus)] and slow (soleus) fibres, and skeletal muscle SR subfractions enriched in plasma membrane, terminal cisternae and LSR were prepared as described previously [1,23,24]. In order to determine whether SRP-35 is an integral membrane protein, mouse skeletal muscle SR vesicles were treated with Na2CO3/KCl as described previously [25]. Mouse skeletal muscle SR (15 μg) were digested with increasing concentrations of trypsin for 2 min at room temperature. The reaction was blocked by the addition of trypsin inhibitor; samples were then loaded on to SDS/PAGE (12.5% gels), blotted on to nitrocellulose membranes and probed with the indicated antibodies. Localization of GFP-tagged recombinant proteins in transfected C2C12 cells was performed 48 h after transfection. Briefly, myotubes were fixed with 3.7% paraformaldehyde (in PBS), mounted in glycerol medium and observed under fluorescent light (excitation wavelength 480 nm, emission wavelength 510 nm) with an inverted fluorescent microscope (Axiovert S100 TV, Carl Zeiss) equipped with a 20× water-immersion FLUAR objective [0.75 NA (numerical aperture)] attached to a Cascade 128+ CCD (charge-coupled-device) camera (Photometrics).

Polyclonal antibody production, Western blotting and confocal microscopy

Polyclonal antibodies raised against the recombinant GST–SRP-35 fusion protein were obtained by immunizing rabbits and the IgG fraction was purified as described previously [25]. Alternatively, commercial rabbit anti-DHRS7C antibodies were tested on blotted proteins and indirect immuno-enzymatic staining was carried out as described previously [25]. Confocal microscopy was performed on paraffin-embedded mouse skeletal muscle tissue sections; briefly, sections were re-hydrated by incubating sequentially in Xylene (twice, 2 min), 100% ethanol (twice, 30 s), 95% and 80% ethanol (once, 30 s) followed by water (3 times, 3 min) and equilibration in PBS; membranes were permeabilized for 30 min at room temperature with a solution containing 0.5% Triton X-100, 2% (v/v) horse serum and 1% (v/v) BSA; slides were then incubated with the primary antibody (rabbit anti-DHRS7C and mouse anti-SERCA monoclonal antibody) diluted in 0.01% Triton X-100, 1% BSA and 2% (v/v) horse serum in PBS overnight at 4°C, followed by 4 washes of 15 min each with PBS and incubation for 40 min at room temperature with Alexa Fluor®-488 chicken anti-rabbit IgG and Alexa Fluor®-568 donkey anti-mouse IgG diluted in 0.01% Triton X-100, 1% (v/v) BSA and 2% (v/v) horse serum in PBS. Slides were mounted and fluorescence was observed by confocal microscopy using a Leica DM1400 confocal microscope equipped with a 100× HCX APO TIRF objective (1.47 NA).

Affinity chromatography

In order to determine whether SRP-35 is a dinucleotide-binding protein, we used the affinity column HiTrap Blue Sepharose; the ligand Cibacron Blue F3G-A shows structural similarities to NAD(H), enabling proteins that bind strongly to this cofactor to attach to the resin [26]. Total SR vesicles (1 mg/ml) were solubilized for 30 min at 4°C in a solution containing 10 mM Tris/HCl, pH 8.0, 1% (v/v) DDM (N-dodecyl-β-D-maltoside) and 1 M NaCl. The solubilized SR membrane proteins were diluted 10-fold with a solution containing Tris/HCl, pH 8.0, and 1 M NaCl and then incubated with Cibacron Blue F3G-A Sepharose previously equilibrated with 10 mM Tris/HCl, pH 8.0, 0.1% DDM and 1M NaCl. The resin was washed with 10 bed volumes of buffer containing 10 mM Tris/HCl, pH 8.0, 0.1% DDM and 1M NaCl, and proteins were eluted by adding 1 mM NADH. In order to verify if some SRP-35 was still bound to the resin, 100 μl of Laemmli loading buffer were added to the Cibacron Blue F3G-A Sepharose, after 5 min of incubation, samples were centrifuged and 30 μl of supernatant were loaded on the gel.

Ca2+ release and KCl dose–response curves

C2C12 cells, that were grown on glass cover slips, transfected (with pEYFPC1 or pEYFPC1-SRP-35) and differentiated, were loaded with the ratiometric Ca2+ indicator fura 2 acetoxymethylester (5 μM) in Krebs–Ringer solution (140 mM NaCl, 5 mM KCl, 1 mM Mg2+, 20 mM Hepes, 1 mM NaHPO4, 5.5 mM glucose, when indicated, and 2 mM Ca2+, pH 7.4) for 30 min at 37°C. YFP-positive cells were first identified using a 40× Plan-Neofluar objective (NA 1.3) and filter set N°44 (Carl Zeiss MicroImaging; BP 475/40, FT 500, BP 530/50) as described previously [24]. Transfected cells were then analysed for their response to pharmacological activation with KCl or 4-chloro-m-cresol, or for the status of the intracellular Ca2+ stores, by treating the cells with 5 μM ionomycin plus 1 μM thapsigargin in Krebs–Ringer solution containing no added Ca2+ and 0.5 mM EGTA, by monitoring the changes in fura 2 fluorescence [27]. Images were acquired with a Cascade 128+ CCD camera and analysed using the Metamorph imaging software version 7.7.0.0. Generation of dose–response curves and statistical analysis were performed using GraphPad Prism version 4.00 (GraphPad Software). For some experiments, myotubes were treated with retinoic acid (5 μM) for 4 days, loaded with fura 2 and analysed for their calcium response to KCl and 4-chloro-m-cresol as indicated above.

Retinol/retinal enzymatic assays

HEK-293 cells were transfected with pEYFPC1 or pEYFPC1-SRP-35, and 24 h post-transfection either all-trans-retinol (10 μM) was added to the cell medium for 6 h, or all-trans-retinaldehyde (5 μM) was added to the cell medium for 3 h. After incubation, the cells were washed twice with PBS and the medium was replaced with 1 ml 100% methanol and 1 ml 2 M hydroxylamine, pH 6.7. Cells were subsequently scraped off from the tissue culture flask, homogenized in a glass potter and after a 10 min incubation at room temperature, the homogenate was stored at −80°C until used.

For retinol quantification, all solutions were made diluting a stock solution of all-trans-retinol or all-trans-retinaldehyde prepared in 100% ethanol, to an aqueous solution containing 10 μM BSA (when all-trans-retinol was added) or 5 μM BSA (when all-trans-retinaldehyde was added) followed by sonication (23°C, 10 min continuous pulse at maximum setting using a Branson sonifier, 2210E-MTH 47 KHz/234 W). The final concentration of ethanol did not exceed 1% (v/v). This procedure guarantees a more soluble and stable retinoid substrate solution [28]. To quench the reaction, 1 vol. of 2 M hydroxylamine, pH 6.7, and 1 vol. of 100% methanol were added, which also stabilized the retinoids. Retinoid extraction from 500 μl of the homogenates was optimized by adding acetone (1 vol.) as a phase-mixing agent followed by three times extraction with petrol ether (0.6 vol.). The organic solvent was evaporated under a stream of nitrogen and the retinoid pellet was dissolved in 100 μl of HPLC mobile phase (99.5:0.5 hexane/ethanol), and injected into the HPLC. The relative amounts of retinaldehyde and retinol are expressed as percentages (the amount of a retinoid species divided by the total amount of all retinoids present).

Real-time PCR

Total RNA was extracted from C2C12 cells and treated with deoxyribonuclease I as described previously [29]. After RT using 1000 ng of RNA, cDNA was amplified by quantitative real-time PCR using SYBR Green technology as described previously [29] and the following primers: RyR1, forward 5′-GCACACAGTCGTATGTACCTG-3′ and reverse 5′-CCTCCCCTGTTGCGTCTTC-3′; SRP-35, forward 5′-CCCTGGAGCTTGACAAAAAGA-3′ and reverse 5′-GTTCACTAACACAATCTGGCCT-3′; and Cav1.1, forward 5′-TCAGCATCGTGGAATGGAAAC-3′ and reverse 5′-GTTCAGAGTGTTGTTGTCATCCT-3′. Gene expression was normalized using self-TATA-box binding protein as a reference with the following primers: forward 5′-GCCATAAGGCATCATTGGAC-3′ and reverse 5′-AACAACAGCCTGCCACCTTA-3′.

Software and statistical analysis

BLAST alignments were performed on the NCBI (National Center for Biotechnology Information) web site using BLAST 2.2.8. Multiple sequence alignments were performed using the ClustalW algorithm available from the Swiss node of the European Molecular Biology Network. Statistical analysis was performed using the Student's t test for two populations. Values were considered significant when P<0.05.

RESULTS

Primary structure of SRP-35

The skeletal muscle SR junctional face membrane is enriched in proteins playing a major role in Ca2+ homoeostasis. One of the aims of the research of our and other laboratories is to identify all the protein constituents of this membrane fraction [10,11,30]. Results obtained by electrospray MS analysis revealed the presence of a novel polypeptide with an approximate molecular mass of 35 kDa in the SR. On the basis of the amino acid sequence obtained from two peptides of the 35 kDa rabbit skeletal muscle protein (Figure 1, light grey boxes), primers were designed and used to amplify by RT–PCR (from mouse skeletal muscle mRNA) a cDNA sequence of approximately 960 nucleotides, whose primary sequence matched that of a hypothetical mouse protein present in the NCBI database (see Figure 1). The predicted primary sequence of SRP-35 encompasses: (i) the peptide sequences obtained from the rabbit skeletal muscle protein from which primers were designed (light grey boxes); (ii) an N-terminal hydrophobic sequence (dark grey box) which may be a signal sequence or a transmembrane domain; and (iii) an NAD(P)(H)-binding and catalytic site (underlined sequence) - the bold letters indicate perfectly conserved residues within the NAD(P)(H)-binding and catalytic sites. BLAST search analysis of the human genome revealed that the human homologue of the rabbit protein maps to human chromosome 17p13.1 (GenBank® accession number NM_001105571) and specifically to the DRS7C_human locus (dehydrogenase/reductase SDR family member 7C or SDR32C2); polymorphisms/mutations have so far not been associated with any human genetic disorder. Figure 1 also shows that SRP-35 is conserved among mammals (87–95% identity with the mouse sequence); however, not all vertebrates express the 7C member of the SDR family of dehydrogenase/reductases, and in fact BLAST analysis revealed that the gene encoding SDR7C is less conserved but present in the genomes of Xenopus (73% identity), Drosophila (38% identity) and zebrafish (58% identity).

Predicted primary sequence of mouse SRP-35 and comparison with that of other vertebrates

Figure 1
Predicted primary sequence of mouse SRP-35 and comparison with that of other vertebrates

Dark grey boxes indicate hydrophobic residues encoding the putative transmembrane domain. The two light grey boxes indicate the peptide sequences obtained from rabbit skeletal muscle SRP-35 from which primers were designed. Underlined residues indicate the catalytic consensus sequence and the bold letters indicate perfectly conserved residues within the NAD(P)(H)-binding site.

Figure 1
Predicted primary sequence of mouse SRP-35 and comparison with that of other vertebrates

Dark grey boxes indicate hydrophobic residues encoding the putative transmembrane domain. The two light grey boxes indicate the peptide sequences obtained from rabbit skeletal muscle SRP-35 from which primers were designed. Underlined residues indicate the catalytic consensus sequence and the bold letters indicate perfectly conserved residues within the NAD(P)(H)-binding site.

Tissue distribution, subcellular localization and membrane topology of SRP-35

In order to gain information concerning its physiological function, we analysed the tissue distribution of SRP-35 by Western blot using two anti-SRP35 polyclonal antibodies (with similar results), one raised against the GST–SRP-35 fusion protein and the other, commercially available, antibody raised against the putative DHRS7C gene product. Figure 2(A) shows that an immunoreactive band of approximately 35 kDa is present in the total homogenate of different mouse tissues, including adipose tissue, liver and skeletal muscle. Interestingly, SRP-35 is not present in the microsomal fraction of the liver, but remains in the 100000 g supernatant (Figure 2B), whereas in skeletal muscle it is highly enriched in the total SR fraction (Figure 2C) and undetectable in the post-microsomal supernatant (results not shown). In all other tissues tested, the 35 kDa immunoreactive protein is undetectable in the microsomal fraction. These results indicate that in muscle, SRP-35 may interact with macromolecules that are exclusively present in skeletal muscle. As we were interested in the functional role of SRP-35 in skeletal muscle, all subsequent experiments were performed on this tissue.

Tissue and subcellular distribution of SRP-35

Figure 2
Tissue and subcellular distribution of SRP-35

(A) Total homogenates (30 μg/lane) from the indicated mouse tissues were separated on SDS/PAGE (12.5% gel), blotted on to nitrocellulose membranes and probed with anti-SRP-35 antibody. (B) Total liver homogenate, the microsomal fraction (100000 g pellet) or post-microsomal supernatant (100000 g supernatant) (30 μg/lane) were loaded on to SDS/PAGE (12.5% gel), blotted on to nitrocellulose membranes and probed with anti-SRP-35 antibody. (C) Total SR from mouse skeletal muscle or the microsomal fraction from the indicated tissues (50 μg/lane) were separated on SDS/PAGE (10% gel), blotted on to nitrocellulose membranes and probed with anti-SRP-35 antibody. (D) Total mouse SR was fractionated into light (PM), LSR and terminal cisternae (TC) (30 μg/lane); proteins were separated on SDS/PAGE (10% gel), blotted on to nitrocellulose and probed with anti-SRP-35 antibody. (E) C2C12 myotubes transfected with the plasmid encoding pEGFP were visualized by brightfield (left) or under fluorescent light with a 20× water-immersion FLUAR objective (0.75 NA). (F) C2C12 myotubes transfected with the plasmid encoding SRP-35–EGFP were visualized by brightfield (left) or under fluorescent light. Note that in these cells fluorescence is punctuated and excluded from the nuclei. Scale bar indicates 50 μm. (G) Longitudinal sections of mouse skeletal muscle were stained with anti-SRP-35 antibody followed by anti-rabbit-conjugated Alexa-Fluor® 488 antibody. Muscle tissue was visualized with a Leica DM1400 confocal microscope equipped with a HCX APO 100× oil immersion TIRF objective (1.47 NA). Images (1024×1024) were acquired at 400 Hz through a 1 μm pinhole; note the striated distribution of SRP-35 compatible with a membrane localization. Scale bar indicates 10 μm. (H) Total homogenates (20 and 40 μg) of fast (EDL) and slow (soleus) muscles were loaded on to SDS/PAGE (10% gel), the proteins were blotted on to nitrocellulose membranes and probed with anti-SRP-35 antibody as described above. Immunoreactivity with albumin was used as a control to show that similar amounts of proteins were loaded. Molecular masses in kDa are shown to the left-hand side of the Western blots in (A)–(D)

Figure 2
Tissue and subcellular distribution of SRP-35

(A) Total homogenates (30 μg/lane) from the indicated mouse tissues were separated on SDS/PAGE (12.5% gel), blotted on to nitrocellulose membranes and probed with anti-SRP-35 antibody. (B) Total liver homogenate, the microsomal fraction (100000 g pellet) or post-microsomal supernatant (100000 g supernatant) (30 μg/lane) were loaded on to SDS/PAGE (12.5% gel), blotted on to nitrocellulose membranes and probed with anti-SRP-35 antibody. (C) Total SR from mouse skeletal muscle or the microsomal fraction from the indicated tissues (50 μg/lane) were separated on SDS/PAGE (10% gel), blotted on to nitrocellulose membranes and probed with anti-SRP-35 antibody. (D) Total mouse SR was fractionated into light (PM), LSR and terminal cisternae (TC) (30 μg/lane); proteins were separated on SDS/PAGE (10% gel), blotted on to nitrocellulose and probed with anti-SRP-35 antibody. (E) C2C12 myotubes transfected with the plasmid encoding pEGFP were visualized by brightfield (left) or under fluorescent light with a 20× water-immersion FLUAR objective (0.75 NA). (F) C2C12 myotubes transfected with the plasmid encoding SRP-35–EGFP were visualized by brightfield (left) or under fluorescent light. Note that in these cells fluorescence is punctuated and excluded from the nuclei. Scale bar indicates 50 μm. (G) Longitudinal sections of mouse skeletal muscle were stained with anti-SRP-35 antibody followed by anti-rabbit-conjugated Alexa-Fluor® 488 antibody. Muscle tissue was visualized with a Leica DM1400 confocal microscope equipped with a HCX APO 100× oil immersion TIRF objective (1.47 NA). Images (1024×1024) were acquired at 400 Hz through a 1 μm pinhole; note the striated distribution of SRP-35 compatible with a membrane localization. Scale bar indicates 10 μm. (H) Total homogenates (20 and 40 μg) of fast (EDL) and slow (soleus) muscles were loaded on to SDS/PAGE (10% gel), the proteins were blotted on to nitrocellulose membranes and probed with anti-SRP-35 antibody as described above. Immunoreactivity with albumin was used as a control to show that similar amounts of proteins were loaded. Molecular masses in kDa are shown to the left-hand side of the Western blots in (A)–(D)

The subcellular distribution of SRP-35 was analysed: (i) in isolated fractions of the SR/ER (endoplasmic reticulum); (ii) by monitoring the expression of GFP-tagged SRP-35 in C2C12 myotubes; and (iii) by confocal microscopy on mature mouse skeletal muscle longitudinal sections. Figure 2(D) shows that SRP-35 is enriched in the skeletal muscle fraction corresponding to the LSR. Quantitative immunoblot analysis revealed that the immunopositive band corresponding to SRP-35 is enriched 6-fold (5.97±0.2, mean±S.E.M.; n=6) in LSR compared with total muscle homogenate (see Supplementary Figure S1 at http://www.BiochemJ.org/bj/441/bj4410731add.htm). These results are compatible with the distribution of fluorescence in C2C12 myotubes transfected with GFP-tagged SRP-35 (Figure 2F). As shown, transfection of C2C12 cells with pEGFP results in an even cytoplasmic distribution of GFP fluorescence, whereas when cells expressed the SRP-35–GFP fusion protein, the resulting punctuated fluorescence is compatible with the localization of SRP-35 in membrane-bound organelles. High-resolution confocal analysis on longitudinal sections of mouse skeletal muscle, shows a cross-striated distribution of endogenous SRP-35 with a centre to centre distance between the bands of approximately 1.5 μm. This fluorescent pattern partially overlaps with that of the SERCA pump (see Supplementary Figure S2 at http://www.BiochemJ.org/bj/441/bj4410731add.htm), indicating a common subcellular distribution. A similar expression pattern was observed after in vivo transfection of FDB (flexor digitorum brevis) fibres with the SRP-35–EGFP construct (results not shown). Interestingly, and as shown in Figure 2(H), slow twitch fibres contain significantly less SRP-35 than fast twitch fibres (means±S.E.M., n=6, percentage intensity of the immunopositive band in Western blots from soleus was 69.7±5.6% of that from EDL; P<0.0001).

On the basis of its deduced primary sequence, the N-terminal of this protein encompasses a hydrophobic peptide of 24 residues, which could be a signal sequence or may form a trans-membrane α-helical segment. To establish whether in skeletal muscle SRP-35 is an integral membrane protein we: (i) extracted SR vesicles with KCl/Na2CO3; and (ii) performed trypsin digestion followed by Western blot analysis. Treatment of mouse total SR membrane fraction with 0.6M KCl, a procedure used to separate loosely bound proteins from the membrane of microsomal vesicles, did not dissociate SRP-35 from the SR vesicles (Figure 3A, left-hand panel). The failure to extract SRP-35 with high-ionic strength wash might be due to the localization of SRP-35 in the lumen of the SR vesicles, similarly to calsequestrin (Figure 3A, right-hand panel), or it might result from its integral association to the SR membrane. To discriminate between these possibilities, we performed an alkaline extraction of the SR fraction with 100 mM Na2CO3. As expected, calsequestrin (Figure 3B, right-hand panel), an intra-lumenal SR protein, was found in the Na2CO3-solubilized fraction. In contrast, SRP-35 was still associated with the insoluble membrane fraction (Figure 3B, left-hand panel), as was calnexin, a 90 kDa integral membrane protein of the ER (Figure 3B, middle panel). We further tested under what conditions SRP-35 could be solubilized, by treating SR vesicles with different detergents [CHAPS, DHPC (1,2-diheptanoyl-sn-glycero-3-phosphocholine) and DDM] under high ionic conditions. As shown, extraction with 1% (w/v) CHAPS was not efficient, whereas the treatment of SR vesicles with 1% (w/v) DHPC or DDM in the presence of 1 M NaCl was equally efficient at solubilizing SRP-35 (Figure 3C). Finally, the topological orientation of SRP-35 was assessed by proteolysis of SR vesicles with increasing concentrations of trypsin. Under our experimental conditions, the immunoreactivity of SRP-35 decreased with increasing trypsing concentration even at very low trypsin concentrations (30 and 150 ng), whereas at these concentrations the immunoreactivity of calregulin, which is a luminal SR protein and thus not exposed to trypsin, was unaffected (Figure 3D). These results indicate that the majority of the SRP-35 polypeptide, including its epitope(s), protrudes from the SR vesicles and is exposed towards the myoplasm and only a small portion is bound to the SR membrane.

SRP-35 is tightly associated with SR membranes and the majority of the protein faces the myoplasm

Figure 3
SRP-35 is tightly associated with SR membranes and the majority of the protein faces the myoplasm

Total mouse SR microsomes were incubated for 30 min on ice with either KCl (A) or Na2CO3 (B) and centrifuged at 150000 g for 45 min. Proteins present in the supernatant and in the pellet were separated on SDS/PAGE (12.5% gels) and stained with anti-SRP-35 antibodies or Stains-All (to visualize calsequestrin which stains metachromatically blue). (C) Total mouse skeletal muscle SR (1 mg/ml) was treated with the indicated detergent (1%) in the presence of 1 M NaCl for 30 min at 4°C. After centrifugation at 135000 g, pellets and supernatant were collected and proteins (30 μg/lane) were separated by SDS/PAGE (12.5% gels), blotted on to nitrocellulose membranes and stained with anti-SRP-35 antibody. (D) Total SR from mouse skeletal muscles (50 μg of protein) were treated with increasing concentrations of trypsin for 2 min at room temperature. After treatment, the reaction was blocked and proteins were loaded on to SDS/PAGE (12.5%), blotted on to nitrocellulose membranes and stained with antibodies against SRP-35 (left) and calregulin (as a control for vesicle integrity). NT, no treatment. Molecular masses in kDa are shown to the left-hand side of the Western blots.

Figure 3
SRP-35 is tightly associated with SR membranes and the majority of the protein faces the myoplasm

Total mouse SR microsomes were incubated for 30 min on ice with either KCl (A) or Na2CO3 (B) and centrifuged at 150000 g for 45 min. Proteins present in the supernatant and in the pellet were separated on SDS/PAGE (12.5% gels) and stained with anti-SRP-35 antibodies or Stains-All (to visualize calsequestrin which stains metachromatically blue). (C) Total mouse skeletal muscle SR (1 mg/ml) was treated with the indicated detergent (1%) in the presence of 1 M NaCl for 30 min at 4°C. After centrifugation at 135000 g, pellets and supernatant were collected and proteins (30 μg/lane) were separated by SDS/PAGE (12.5% gels), blotted on to nitrocellulose membranes and stained with anti-SRP-35 antibody. (D) Total SR from mouse skeletal muscles (50 μg of protein) were treated with increasing concentrations of trypsin for 2 min at room temperature. After treatment, the reaction was blocked and proteins were loaded on to SDS/PAGE (12.5%), blotted on to nitrocellulose membranes and stained with antibodies against SRP-35 (left) and calregulin (as a control for vesicle integrity). NT, no treatment. Molecular masses in kDa are shown to the left-hand side of the Western blots.

SRP-35 binds the dinucleotide NAD(H) and has catalytic activity

In silico comparison analysis of the deduced amino acid sequence of SRP-35 revealed that it shares homology with proteins belonging to the SDR family. In order to determine whether it indeed binds dinucleotides, the affinity column HiTrap Blue Sepharose was used; the ligand Cibacron Blue F3G-A shows structural similarities to NAD(H), enabling proteins strongly binding to this cofactor to attach to the resin [26]. Solubilized SR proteins were incubated with HiTrap Blue Sepahrose, the resin was extensively washed and bound proteins were eluted with 1 M NaCl plus NADH. The fraction(s) enriched in SRP-35 were identified by immunostaining with anti-SRP-35 antibodies. Figure 4 shows a representative result; the left-hand panels of Figures 4(A) and 4(B) are Ponceau-Red-stained membranes, whereas the right-hand panels are the same blots stained with anti-SRP-35 antibodies (Figure 4A) and with anti-calsequestrin antibodies as a control (Figure 4B). An immunoreactive band corresponding to SRP-35 is present in the total SR (lane 1), in the solubilized SR fraction (lane 3) and in the NADH-eluted fraction (lane 6). Not all SRP-35 was eluted with NADH, as demonstrated by the presence of an immunoreactive band of 35 kDa in an aliquot of the Cibacron Blue F3G-A resin (lane 7). On the other hand, when the same experiment was performed by following the distribution of calsequestrin, a major SR protein which should not bind to Cibacron Blue F3G-A, its immunoreactivity is present in the total SR, but not in the NADH eluate nor bound to the resin.

SRP-35 is a dinucleotide-binding protein and its overexpression leads to an increased production of all-trans-retinaldehyde

Figure 4
SRP-35 is a dinucleotide-binding protein and its overexpression leads to an increased production of all-trans-retinaldehyde

SRP-35 binds Cibacron Blue F3G-A ligand. (A) 1 mg/ml total SR vescicles were solubilized in a solution containing 1% DDM, 1 M NaCl and subsequently incubated with Cibacron Blue F3G-A resin as described in the Experimental section. Proteins present in the total SR (30 μg, lane 1), the pellet after solubilization with DDM/NaCl (30 μl, lane 2), total SR proteins solubilized with DDM/NaCl (30 μl, lane 3), void (30 μl, lane 4), last wash (30 μl, lane 5), eluted with NADH (30 μl, lane 6) and attached to the resin (30 μl, lane 7). Left-hand panel, Ponceau Red-stained blot; right-hand panel, blot stained with anti-SRP-35 Abs. (B) as in (A), except that the right-hand blot was stained with anti-calsequestin antibodies. Note that calsequestrin does not bind to Cibacron Blue F3G-A. (C and D) Chromograms (325 nm) of the homogenates of pEYFPC1- and pEYFPC1-SRP-35-transfected HEK-293 cells. All-trans-retinylester (1), all-trans-retinaldehyde-oxim (syn: 2; anti: 5) (RAL), 13-cis-retinol (3) and all-trans retinol (4) (ROL). The peaks were identified by authentic standards. (E and F) Bar histograms comparing the percentage of retinaldehyde (RAL) and retinol (ROL) present in pEYFPC1- and pEYFPC1-SRP-35-transfected cells. Results are the means±S.E.M. of the experiments. The percentage of retinaldehyde in cells overexpressing SRP-35 was significantly higher than in YFP-transfected cells, P<0.019. Molecular masses in kDa are shown to the left-hand side of the Western blots in (A) and (B).

Figure 4
SRP-35 is a dinucleotide-binding protein and its overexpression leads to an increased production of all-trans-retinaldehyde

SRP-35 binds Cibacron Blue F3G-A ligand. (A) 1 mg/ml total SR vescicles were solubilized in a solution containing 1% DDM, 1 M NaCl and subsequently incubated with Cibacron Blue F3G-A resin as described in the Experimental section. Proteins present in the total SR (30 μg, lane 1), the pellet after solubilization with DDM/NaCl (30 μl, lane 2), total SR proteins solubilized with DDM/NaCl (30 μl, lane 3), void (30 μl, lane 4), last wash (30 μl, lane 5), eluted with NADH (30 μl, lane 6) and attached to the resin (30 μl, lane 7). Left-hand panel, Ponceau Red-stained blot; right-hand panel, blot stained with anti-SRP-35 Abs. (B) as in (A), except that the right-hand blot was stained with anti-calsequestin antibodies. Note that calsequestrin does not bind to Cibacron Blue F3G-A. (C and D) Chromograms (325 nm) of the homogenates of pEYFPC1- and pEYFPC1-SRP-35-transfected HEK-293 cells. All-trans-retinylester (1), all-trans-retinaldehyde-oxim (syn: 2; anti: 5) (RAL), 13-cis-retinol (3) and all-trans retinol (4) (ROL). The peaks were identified by authentic standards. (E and F) Bar histograms comparing the percentage of retinaldehyde (RAL) and retinol (ROL) present in pEYFPC1- and pEYFPC1-SRP-35-transfected cells. Results are the means±S.E.M. of the experiments. The percentage of retinaldehyde in cells overexpressing SRP-35 was significantly higher than in YFP-transfected cells, P<0.019. Molecular masses in kDa are shown to the left-hand side of the Western blots in (A) and (B).

Since SRP-35 shares homology with the catalytic domains of classical SDR family members that convert, among others, retinol substrates to retinaldehyde derivatives, concomitantly reducing NAD, we set up an assay to validate the enzymatic activity of SRP-35 by verifying if its over-expression affects the amount of all-trans-retinol converted to all-trans-retinaldehyde. As an experimental setup we chose an ‘in situ’ whole cell assay in which transiently transfected cells (HEK-293 cells or C2C12 myotubes) were incubated with all-trans-retinol or all-trans-retinaldehyde, and the amount of product formed was compared with that obtained in cells transiently transfected with the empty pEYFP plasmid. Figures 4(C) and 4(D) show original HPLC chromatograms obtained from lipid extracts of YFP- and SRP-35–YFP-transfected HEK-293 cells, Figures 4(E) and 4(F) show the cumulative results obtained by pooling data from three independent transfection experiments and normalized for the relative amount of all-trans-retinaldehyde-oxim (peaks 2 and 5) and retinol (peak 4) divided by the total amount of peaks 1–5, defined as 100% (the total amount of all intracellular retinoids present in the lipid extracts). As shown, HEK-293 cells overexpressing SRP-35 generate a significantly larger proportion of all-trans-retinaldehyde (sum of peaks 2 and 5) compared with YFP-transfected controls (Figure 4E). Interestingly, the reaction appears to preferentially proceed in the forward oxidative reaction, since no change in the relative amount of all-trans-retinol was obtained in SRP-35 overexpressing cells (Figure 4F). Similar results were obtained in C2C12-transfected myotubes (results not shown). In addition, 13-cis-retinol was sometimes detected in the SRP-35-transfected HEK-293 cells (peak 3), but may be an artefactual change of all-trans-retinol due to the extraction procedure. We also verified if SRP-35 overexpression affects the overall redox status of cells by performing Oxy-Blot analysis on total extracts of GFP- and SRP-35–GFP-transfected HEK-293 cells; as shown in Supplementary Figure S3 (at http://www.BiochemJ.org/bj/441/bj4410731add.htm), cells transfected with the SRP-35 fusion protein exhibited a significant increase in the quantity of oxidatively modified proteins, confirming that SRP-35 is indeed involved in redox reactions.

Effect of SRP-35 overexpression on excitation–contraction coupling

SRP-35 is located within a subcellular membrane devoted to the regulation of the Ca2+ concentration, thus we investigated if its overexpression affects Ca2+ homoeostasis. C2C12 mytotubes were transfected with SRP-35–EYFP or with the empty pEYFP vector, loaded with fura 2 and individually stimulated with different concentrations of either KCl (mimicking depolarization) or 4-chloro-m-cresol (inducing direct activation of the RyR1). As shown in Figure 5, overexpression of SRP-35 affected neither the resting Ca2+ concentration (Figure 5C) nor the EC50 for KCl (26.2±0.8 and 27.6±1.8 mM for YFP and SRP-35–YFP overexpressing cells respectively; Figure 5A) and 4-chloro-m-cresol (340.6±32.6 and 306.7±16.9 μM for YFP and SRP-35–YFP over-expressing cells respectively; Figure 5B) induced Ca2+ release; however, it significantly decreased the peak Ca2+ release induced by maximal stimulatory concentrations of KCl by approximately 40% (Figure 5D; P<0.05) and 4-chloro-m-cresol (Figure 5E; P<0.008). This decrease in peak Ca2+ release was not due to depletion of intracellular stores, since the peak Ca2+ transients elicited by a treatment aimed at depleting intracellular stores (5 μM ionomycin plus 1 μM thapsigargin in 0.5 mM EGTA) were not significantly different (Δ fluorescence ratios were 0.251±0.041 and 0.205±0.048; P=0.479 in YFP- and SRP-35–YFP-expressing cells). Interestingly, addition of retinoic acid to C2C12 myotubes during myotube differentiation mimicked the effect of SRP-35 overexpression by causing a significant decrease in the peak Ca2+ concentration induced by KCl (Figure 5D; P<0.01) and 4-chloro-m-cresol (Figure 5E; P<0.008). The reduced Ca2+ release could be due to allosteric regulation by products deriving from the enzymatic activity of SRP-35, such as NADH and/or to alterations of the level of expression of the proteins involved in excitation–contraction coupling. The latter event may result from the effect of retinoic acid generated from the retinaldehyde produced by SRP-35. In order to investigate this, we performed real-time PCR on mRNA extracted from C2C12 myotubes treated in culture with 5 μM retinoic acid. We chose this approach to avoid the variability linked to the efficiency of transient transfections of C2C12 cells. Figure 5(F) shows that retinoic acid treatment of C2C12 myotubes induces a 2.5-fold induction in the level of expression of SRP-35; no effect was observed on the expression level of the α1.1 subunit of the voltage-sensing DHPR (Figure 5G), but, interestingly, it caused a significant reduction (Figure 5H; P<0.025) in the expression level of the mRNA encoding RyR1. This decrease in the RyR1 transcript was paralleled by a decrease in the content of the RyR1 protein as confirmed by Western blot analysis (see Supplementary Figure S4 at http://www.BiochemJ.org/bj/441/bj4410731add.htm). The extent of decrease of the mRNA level is similar to the extent of decrease of Ca2+ release observed after pharmacological activation of the RyR1 and most likely results from changes in the protein level of the RyR1 induced by retinoic acid production after 6 days of SRP-35 over-expression.

Effect of SRP-35 on calcium fluxes in C2C12 myotubes

Figure 5
Effect of SRP-35 on calcium fluxes in C2C12 myotubes

(A and B) KCl- and 4-chloro-m-cresol dose–response curves in C2C12 transfected with pEYFPC1 (full line) or pEYFPC1-SRP-35 (dashed line). Data points represent the mean peak fluorescence increase of 6–11 cells. Curves were fitted using a Boltzmann equation; there were no significant differences in the EC50 for KCl or 4-chloro-m-cresol. (C) SRP-35 over-expression does not affect the resting Ca2+ concentrations of C2C12 cells. (D and E) Peak fura-2 Δ fluorescence (means±S.E.M. of the indicated number of cells) in C2C12 transfected with pEYFPC1 (YFP), pEYFPC1-SRP-35 (SRP-35), untreated (cont) and treated with retinoic acid (RA; 5 μM) for 4 days. *P<0.005; **P<0.01; ***P<0.008. (FH) Real-time RT–PCR on control or retinoic acid-treated C2C12 cells. retinoic acid does not affect the relative expression level of CaV1.1, but it significantly increases the relative expression level of SRP-35 while decreasing that of RyR1. Results are expressed as means±S.E.M. of the indicated number of experiments. ***P<0.001.

Figure 5
Effect of SRP-35 on calcium fluxes in C2C12 myotubes

(A and B) KCl- and 4-chloro-m-cresol dose–response curves in C2C12 transfected with pEYFPC1 (full line) or pEYFPC1-SRP-35 (dashed line). Data points represent the mean peak fluorescence increase of 6–11 cells. Curves were fitted using a Boltzmann equation; there were no significant differences in the EC50 for KCl or 4-chloro-m-cresol. (C) SRP-35 over-expression does not affect the resting Ca2+ concentrations of C2C12 cells. (D and E) Peak fura-2 Δ fluorescence (means±S.E.M. of the indicated number of cells) in C2C12 transfected with pEYFPC1 (YFP), pEYFPC1-SRP-35 (SRP-35), untreated (cont) and treated with retinoic acid (RA; 5 μM) for 4 days. *P<0.005; **P<0.01; ***P<0.008. (FH) Real-time RT–PCR on control or retinoic acid-treated C2C12 cells. retinoic acid does not affect the relative expression level of CaV1.1, but it significantly increases the relative expression level of SRP-35 while decreasing that of RyR1. Results are expressed as means±S.E.M. of the indicated number of experiments. ***P<0.001.

DISCUSSION

The skeletal muscle SR is an organelle that is specialized at regulating calcium homoeostasis, and identifying all its protein components constitutes a major step towards the elucidation of the regulation of calcium under normal and pathological conditions. In the present study we identified SRP-35 and show that this novel component of the skeletal muscle SR belongs to the SDR protein family, an enzyme whose activity may link calcium homoeostasis to activation of metabolism.

The enzyme superfamily of SDRs consits of more than 46000 family members that are ubiquitously expressed and whose transcripts have been found in virtually all genomes investigated [12,31,32]. In the presence of specific co-factors they dehydrogenate numerous substrates, including sterols, 3-hydroxysterols, alcohols, retinols, sugars, aromatic compounds and xenobiotics [33], thus playing critical roles in lipid, amino acid and carbohydrate metabolism; they also regulate many physiological processes by sensing the redox status in metabolism and transcription [13]. Although SDRs share little genetic similarity (15–30%) [32], some sequence motifs in their tertiary structures resemble each other [13] and, importantly, their co-factor-binding domain is highly conserved. In humans and mouse at least 70 distinct SDRs have been identified: they have a highly variable C-terminal substrate-binding domain, but all contain a Rossmann-fold scaffold and bind NAD(P) dinucleotides [13,31]. SRP-35 contains a tyrosine residue with adjacent lysine and serine residues in its active site, but is slightly larger than ‘classical’ SDR, being 311 rather than the classical 250 residues long; thus it should be classified as an ‘extended’ member of the SDR superfamily [13]. Among the many functions catalysed by SDR family members, some are involved in retinol/retinladehyde metabolism. Indeed, retinols (such as vitamin A) that are taken in with the diet are transported at high concentrations via the serum bound to retinol-binding protein and can be taken up by any cell for storage or potential conversion into retinoic acid. Once inside the cell they are converted into retinal via a reversible reaction catalysed by SDR (or alcohol dehydrogenases), first to retinaldehyde with the concomitant generation of NADH and subsequently to retinoic acid by cytosolic aldehyde dehydrogenases (Aldh1a1 to Aldh1a3) [34,35].

The results of the present study support a role for SRP-35 in the conversion of retinol to retinaldehyde and thus to its involvement in the retinoic acid signalling pathway in skeletal muscle. Skeletal muscles express RAR and high levels of RXRγ [15,36,37], receptors that can activate the muscle-specific transcription factors myogenin and myoD [38,39], and treatment of muscle cells in culture with retinoic acid has been shown to stimulate myotube differentiation [39,40].

An intriguing aspect of SRP-35 is its particular tissue and subcellular distribution; in fact, Western blot analysis of different tissues revealed that it is abundantly expressed in liver and kidney, where it is absent from the microsomal fraction, but present in the post-microsomal supernatant. Indeed, SDRs have been found in the cytosol, in organelles such as the ER and mitochondria as well as in low-density peroxisomes [41,42], the latter being too light to be pelleted by a 1 h centrifugation at 100000 g. In order to investigate whether SRP-35 is present in ‘light’ membrane fractions in liver and kidney, a more detailed analysis of rough and smooth ER, peroxisomes and lysosomes should be carried out. Nevertheless, in skeletal muscle, SRP-35 is present in the SR and on the basis of our subcellular fractionation experiments, it is enriched in the LSR, a fraction also enriched in proteins such as the SERCA pump and sarcalumenin (for a review, see [11]). Such a distribution is compatible with the high-resolution confocal immunofluorescence experiments of endogenous SRP-35, which revealed the presence of cross-striated bands, a pattern which is similar and partially overlaps with the distribution of SERCA. We cannot exclude that in skeletal muscle targeting of SRP-35 to the LSR is mediated by its interaction with other proteins. Co-immunoprecipitation experiments did not reveal any interaction with either SERCA1 or SERCA2 (results not shown) and binding to other SR proteins may be of low affinity and thus easily disrupted by the high salt concentration and detergents required to solubilize SRP-35. Interestingly, quantitative analysis of the content of SRP-35 in slow and fast twitch muscles indicates that this protein is enriched in fast twitch muscles; although of potential significance, this may be due to the fact that the relative volume of the SR is almost double in EDL compared with soleus muscles [4345].

The biochemical characterization of this novel muscle SDR strongly support that it is a membrane-bound protein with its catalytic site facing the cytoplasm; in fact: first, it could not be extracted by treating vesicles with high salt or bicarbonate, methods which have been successfully used to extract loosely bound or soluble proteins from the ER/SR [23,24,46]; and secondly, mild detergents such as CHAPS were not efficient at solubilising SRP-35. Thirdly, trypsin digestion revealed that most of the protein faces the myoplasm; thus both its product (retinaldehyde) and NAD(P)H would be released into the myoplasm. The generation of NADH in the myoplasm may have functional signalling relevance; in fact, lactate dehydrogenase requires NADH as reducing power to generate lactate from pyruvate and, although NADH is available in the mitochondria where it is used to generate ATP, there are no transporters for NADH and therefore the reduced co-factor must be regenerated in the cellular compartments where it is consumed [47]. It should also be mentioned that cytosolic NADH has been shown to regulate the activity of the RyR, especially in the heart [48,49]. Although we are aware that SRP-35 is a low abundance SR protein and we do not exactly know the stoichiometric ratio between SRP-35 and RyR1, we speculate the NADH generated by SRP-35 in a microdomain adjacent to SR membranes might allosterically modulate RyR1 activity. We also can not exclude the possibility that SRP-35 indirectly modulates Ca2+ release by affecting the level of expression of the RyR1 protein.

Our results from the present study show that overexpression of SRP-35 in C2C12 myotubes results in a decrease of RyR1-mediated Ca2+ release, a result that could be mimicked by the addition of retinoic acid to the culture medium during differentiation. Indeed, both the Ca2+ release induced by KCl (mimicking depolarization) as well as that induced by direct activation of RyR1 were reduced by approximately 40% with no significant change in the EC50 for either agonist. Such a result is reminiscent of the effect of mutations in the RYR1 gene linked to some forms of central core disease and multiminicore disease; in fact, mytubes from such patients have reduced pharmacologically evoked Ca2+ release, and, depending on the mutation, increased or normal resting Ca2+ concentration and/or reduced expression of RyR1 in muscle [25,5052]. In the case of SRP-35, the decreased Ca2+ release was not accompanied by a change in the resting Ca2+ concentration nor in the Ca2+ present in intracellular stores, suggesting that the effect was not due to a modification of Ca2+ fluxes across surface membranes [53], but probably due to reduced expression of components of the Ca2+ release units. Real-time PCR on retinoic acid-treated cells shows that the level of expression of the Cav1.1 was not affected, but that of the RyR1 was reduced by approximately 50%, a result confirmed by immunoblotting. A similar effect of retinoic acid treatment on the expression of inositol 1,4,5-trisphoshate receptors has been reported [54,55] and is thought to be due to a decrease in the promoter activity in response to retinoic acid.

What is the function of SRP-35 in vivo and what is the physiological activator of this SDR? Skeletal muscle constitutes approximately 40% of the total body mass, accounts for more than 30% of energy expenditure and is the major tissue involved in insulin-dependent glucose uptake. Metabolism is largely regulated by nuclear hormone receptors that function as regulators of transcription and it has been demonstrated that in mice retinoic acid treatment favours mobilization of body fat, increases fatty acid oxidation and decreases body weight [20,56]. Alhough at the moment it is difficult to envisage what activates the enzymatic activity of SRP-35, its strategic subcellular localization in a compartment dedicated to Ca2+ homoeostasis, and the fact that during the conversion of retinol to retinaldehyde it will concomitantly generate NADH, indicates that it may be an important molecule linking Ca2+ mobilization and thus muscle contraction to the activation of metabolic functions. SRP-35 is a dehydrogenase and could reduce the NAD+ generated by lactate dehydrogenase during sustained muscle activity. In fact, the NAD+ that is generated during glycolysis remains in the cytoplasm since the inner mitochondrial membrane lacks a specific transport system [47]. Interestingly, many enzymes involved in the glycolytic pathway adhere to the SR membranes [21]. The reduction of NAD+ by SRP-35 is coupled to the conversion of all-trans-retinaldehyde to retinoic acid, an important activator of GLUT4 (glucose transporter 4) gene transcription [57] and regulator of RyR1 expression (Figure 5). Additionally, a secondary effect of SRP-35 could be mediated by NADH, which has been shown to regulate RyR activity [49]. The development of an experimental model overexpressing SRP-35 in the skeletal muscle will allow us to investigate in greater detail many aspects of this novel SDR, ranging from metabolism to activation of transcription of specific genes.

Abbreviations

     
  • CCD

    charge-coupled-device

  •  
  • DDM

    N-dodecyl-β-D-maltoside

  •  
  • DHPC

    1,2-diheptanoyl-sn-glycero-3-phosphocholine

  •  
  • DHPR

    dihydropyridine receptor

  •  
  • DHRS7C

    dehydrogenase/reductase (short-chain dehydrogenase/reductase family) member 7C

  •  
  • DMEM

    Dulbecco's modified Eagle's medium

  •  
  • EDL

    extensor digitorum longus

  •  
  • EGFP

    enhanced green fluorescent protein

  •  
  • ER

    endoplasmic reticulum

  •  
  • EYFP

    enhanced yellow fluorescent protein

  •  
  • GFP

    green fluorescent protein

  •  
  • HEK-293

    human embryonic kidney 293

  •  
  • LSR

    longitudinal sarcoplasmic reticulum

  •  
  • NA

    numerical aperture

  •  
  • NCBI

    National Center for Biotechnology Information

  •  
  • RAR

    retinoic acid receptor

  •  
  • RT

    reverse transcription

  •  
  • RXR

    retinoid X receptor

  •  
  • RyR

    ryanodine receptor

  •  
  • SDR

    short-chain dehydrogenase/reductase

  •  
  • SERCA

    sarcoplasmic/endoplasmic reticulum Ca2+-ATPase

  •  
  • SR

    sarcoplasmic reticulum

  •  
  • YFP

    yellow fluorescent protein

AUTHOR CONTRIBUTION

Susan Treves devised the experiments, co-wrote the manuscript, performed the calcium measurements and performed confocal microscopy experiments. Raphael Thurnheer performed the retinol/retinal measurements, bioinformatics, transfection of C2C12 cells and HEK-293 cells and helped perform the calcium imaging experiments. Barbara Mosca, Mirko Vukcevic and Leda Bergamelli created the plasmid constructs, and performed the RT–PCR and molecular biology. Rebecca Voltan performed the original PCR to obtain mouse cDNA, Western blotting, proteolysis, vesicle solubilization and Cibachron Blue chromatography. Vitus Oberhauser supervised and helped in the retinol/retinaldehyde quantification. Michel Ronjat performed the original MS and protein sequencing. Lazslo Csenoch and Peter Szentesi performed the GFP–SRP-35 fluorescence measurements in intact fibres. Francesco Zorzato devised the experiments, supervised all the steps and wrote the paper.

We would like to thank Ms Anne-Sylvie Monnet for her expert technical assistance and Caroline Steiblin for her help with the bioinformatics.

FUNDING

This work was supported by the Department of Anesthesia, Basel University Hospital and by the Association Française contre les Myopathies, Telethon GGP08020, Ministero della Ricerca Scientifica e Tecnologica ex 40 % e 60 % [grant number HPRN-CT-2002-00331].

References

References
1
Saito
A.
Seiler
S.
Chu
A.
Fleischer
S.
Preparation and morphology of sarcoplasmic reticulum terminal cisternae from rabbit skeletal muscle
J. Cell. Biol.
1984
, vol. 
99
 (pg. 
875
-
885
)
2
Schneider
M. F.
Chandler
W. K.
Voltage dependent charge movement of skeletal muscle: a possible step in excitation-contraction coupling
Nature
1972
, vol. 
242
 (pg. 
244
-
246
)
3
Melzer
W.
Hermann-Frank
A.
Luttgau
H. C.
The role of Ca2+ ions in excitation-contraction coupling of skeletal muscle fibres
Biochim. Biophys. Acta
1995
, vol. 
1241
 (pg. 
59
-
116
)
4
Inesi
G.
Canitlina
T.
Yu
X.
Nikic
D.
Sagara
Y.
Kirtley
M. E.
Long-range intramolecular linked functions in activation and inhibition of SERCA ATPases
Ann. N.Y. Acad. Sci.
1992
, vol. 
671
 (pg. 
32
-
47
)
5
MacLennan
D. H.
Ca2+ signaling and muscle disease
Eur. J. Biochem.
2000
, vol. 
267
 (pg. 
5291
-
5297
)
6
MacLennan
D. H.
Asahi
M.
Tupling
A. R.
The regulation of SERCA-type pumps by phospholamban and sarcolipin
Ann. N.Y. Acad. Sci.
2003
, vol. 
986
 (pg. 
472
-
480
)
7
Franzini-Armstrong
C.
Jorgensen
A. O.
Structure and development of E-C coupling units in skeletal muscle
Annu. Rev. Physiol.
1994
, vol. 
56
 (pg. 
509
-
534
)
8
Mitchell
R. D.
Saito
A.
Palade
P.
Fleischer
S.
Morphology of isolated triads
J. Cell Biol.
1983
, vol. 
96
 (pg. 
1017
-
1029
)
9
Rios
E.
Pizarro
G.
Voltage sensor of excitation-contraction coupling in skeletal muscle
Physiol. Rev.
1991
, vol. 
71
 (pg. 
849
-
908
)
10
Weisleder
N.
Takeshima
H.
Ma
J.
Immuno-proteomic approach to excitation-contraction coupling in skeletal and cardiac muscle: molecular insights revealed by the mitsugumins
Cell Calcium
2008
, vol. 
43
 (pg. 
1
-
8
)
11
Treves
S.
Vukcevic
M
Maj
M.
Thurnheer
R.
Mosca
B.
Zorzato
F.
Minor sarcoplasmic reticulum membrane components that modulate excitation-contraction coupling in striated muscles
J. Physiol.
2009
, vol. 
587
 (pg. 
3071
-
3079
)
12
Persson
B.
Kallberg
Y.
Bray
J. E.
Bruford
E.
Dellaporta
S. L.
Favia
A. D.
Duarte
R. G.
Jörnvall
H.
Kavanagh
K. L.
Kedishvili
N.
, et al. 
The SDR (short-chain dehydrogenase/reductase and related enzymes) nomenclature initiative
Chem. Biol. Interact.
2009
, vol. 
178
 (pg. 
94
-
98
)
13
Kavanagh
K. L.
Jörnvall
H.
Persson
B.
Oppermann
U.
Medium- and short-chain dehydrogenase/reductase gene and protein families: the SDR superfamily: functional and structural diversity within a family of metabolic and regulatory enzymes
Cell Mol. Life Sci.
2008
, vol. 
65
 (pg. 
3895
-
3990
)
14
Theodosiou
M.
Laudet
V.
Schubert
M.
From carrot to clinic: an overview of the retinoic acid signaling pathway
Cell Mol. Life Sci.
2010
, vol. 
67
 (pg. 
1423
-
1445
)
15
Mangelsdorf
D. J.
Borgmeyer
U.
Heyman
R. A.
Zhou
J. Y.
Ong
E. S.
Oro
A. E.
Kakizuka
A.
Evans
R. M.
Characterization of three RXR genes that mediate the action of 9-cis retinoic acid
Genes Dev.
1992
, vol. 
6
 (pg. 
329
-
344
)
16
Petkovich
M.
Brand
N. J.
Krust
A.
Chambon
P.
A human retinoic acid receptor which belongs to the family of nuclear receptors
Nature
1987
, vol. 
330
 (pg. 
444
-
450
)
17
Germain
P.
Chambon
P.
Eichele
G.
Evans
R. M.
Lazar
M. A.
Leid
M.
De Lera
A. R.
Lotan
R.
Mangelsdorf
D. J.
Groenmeyer
H.
International Union of Pharmacology. LX. Retinoic acid receptors
Pharmacol. Rev.
2006
, vol. 
58
 (pg. 
712
-
725
)
18
Ziouzenkova
O.
Plutzky
J.
Retinoid metabolism and nuclear receptor responses: new insights into coordinated regulation of the PPAR-RXR complex
FEBS Lett.
2008
, vol. 
582
 (pg. 
32
-
38
)
19
Napoli
J. L.
Biochemical pathways of retinoid transport, metabolism and signal transduction
Clin. Immunol. Immunopathol.
1996
, vol. 
80
 (pg. 
S52
-
S62
)
20
Amengual
J.
Ribot
J.
Bonet
M. L.
Palou
A.
Retinoic acid treatment increases lipid oxidation capacity in skeletal muscle of mice
Obesity
2008
, vol. 
16
 (pg. 
585
-
591
)
21
Xu
K. Y.
Becker
L. C.
Ultrastructural localization of glycolytic enzymes on sarcoplasmic reticulum vesicles
J. Histochem. Cytochem.
1998
, vol. 
46
 (pg. 
419
-
427
)
22
Journet
A.
Chapel
A.
Kieffer
S.
Louwagie
M.
Luche
S.
Garin
J.
Towards a human repertoire of monocytic lysosomal proteins
Electrophoresis
2000
, vol. 
21
 (pg. 
3411
-
3419
)
23
Anderson
A. A.
Treves
S.
Biral
D.
Betto
R.
Sandonà
D.
Ronjat
M.
Zorzato
F.
The novel skeletal muscle sarcoplasmic reticulum JP-45 protein. Molecular cloning, tissue distribution, developmental expression, and interaction with α1.1 subunit of the voltage-gated calcium channel
J. Biol. Chem.
2003
, vol. 
278
 (pg. 
39987
-
39992
)
24
Treves
S.
Feriotto
G.
Moccagatta
L.
Gambari
R.
Zorzato
F.
Molecular cloning, expression, functional characterization, chromosomal localization, and gene structure of junctate, a novel integral calcium binding protein of sarco(endo)plasmic reticulum membrane
J. Biol. Chem.
2000
, vol. 
275
 (pg. 
39555
-
39568
)
25
Zorzato
F.
Anderson
A. A.
Ohlendieck
K.
Froemming
G.
Guerrini
R.
Treves
S.
Identification of a novel 45 kDa protein (JP-45) from rabbit sarcoplasmic-reticulum junctional-face membrane
Biochem. J.
2000
, vol. 
351
 (pg. 
537
-
543
)
26
Sharkis
D. H.
Swenson
R. P.
Purification by cibacron blue F3GA dye affinity chromatography and comparison of NAD(P)H: quinone reductase (E.C.1.6.99.2) from rat liver cytosol and microsomes
Biochem. Biophys. Res. Commun.
1989
, vol. 
161
 (pg. 
434
-
441
)
27
Ducreux
S.
Zorzato
F.
Müller
C.
Sewry
C.
Muntoni
F.
Quinlivan
R.
Restagno
G.
Girard
T.
Treves
S.
Effect of ryanodine receptor mutations on interleukin-6 release and intracellular calcium homeostasis in human myotubes from malignant hyperthermia-susceptible individuals and patients affected by central core disease
J. Biol. Chem.
2004
, vol. 
279
 (pg. 
43838
-
43846
)
28
Gough
W. H.
VanOoteghem
S.
Sint
T.
Kedishvill
N. Y.
cDNA cloning and characterization of a new human microsomal NAD+-dependent dehydrogenase that oxidizes all-trans-retinol and 3 α-hydroxysteroids
J. Biol. Chem.
1998
, vol. 
273
 (pg. 
19778
-
19785
)
29
Treves
S.
Vukcevic
M.
Jeannet
P. Y.
Levano
S.
Girard
T.
Urwyler
A.
Fischer
D.
Voit
T.
Jungbluth
H.
Lillis
S.
, et al. 
Enhanced excitation coupled Ca2+ entry induces nuclear translocation of NFAT and contributes to IL-6 release from myotubes from patients with Central core disease
Hum. Mol. Genetics
2011
, vol. 
20
 (pg. 
589
-
600
)
30
Takeshima
H.
Komazaki
S.
Nishi
M.
Iino
M.
Kangawa
K.
Junctophilins: a novel family of junctional membrane complex proteins
Mol. Cell
2002
, vol. 
6
 (pg. 
11
-
22
)
31
Persson
B.
Krook
M.
Jörnvall
H.
Characteristics of short-chain alcohol dehydrogenases and related enzymes
Eur. J. Biochem.
1991
, vol. 
200
 (pg. 
537
-
543
)
32
Kallberg
Y.
Oppermann
U.
Jörnvall
H.
Persson
B.
Short-chain dehydrogenase/reductase (SDR) relationships: a large family with eight clusters common to human, animal, and plant genomes
Protein Sci.
2002
, vol. 
11
 (pg. 
636
-
641
)
33
Persson
B.
Kallberg
Y.
Oppermann
U.
Jörnvall
H.
Coenzyme-based functional assignments of short-chain dehydrogenases/reductases (SDRs)
Chem. Biol. Interact.
2003
, vol. 
143
 (pg. 
271
-
278
)
34
Duester
G.
Involvement of alcohol dehydrogenase, short-chain dehydrogenase/reductase, aldehyde dehydrogenase, and cytochrome P450 in the control of retinoid signaling by activation of retinoic acid synthesis
Biochemistry
1996
, vol. 
35
 (pg. 
12221
-
12227
)
35
Parés
X.
Farrés
J.
Kedishvili
N.
Duester
G.
Medium- and short-chain dehydrogenase/reductase gene and protein families: the SDR superfamily: functional and structural diversity within a family of metabolic and regulatory enzymes
Cell Mol. Life Sci.
2009
, vol. 
65
 (pg. 
3895
-
3906
)
36
Liu
Q.
Linney
E.
The mouse retinoid-X receptor-γ gene: genomc organization and evidence for functional isoforms
Mol. Endocrinol.
1993
, vol. 
7
 (pg. 
651
-
658
)
37
Smith
A. G.
Muscat
G.E.O.
Skeletal muscle and nuclear hormone receptors: implications for cardiovascular and metabolic disease
Int. J. Biochem. Cell Biol.
2005
, vol. 
37
 (pg. 
2047
-
2063
)
38
Muscat
G.E
Mynett-Johnson
L.
Dowhan
D.
Downes
M.
Griggs
R.
Activation of myoD gene transcription by 3,5,3′-triodo-L-thyronine: a direct role for the thyroid hormone and retinoid X receptors
Nucleic Acids Res.
1994
, vol. 
22
 (pg. 
583
-
591
)
39
Halevy
O.
Lerman
O.
Retinoic acid induces adult muscle cell differentiation mediated by the retinoic acid receptor α
J. Cell Physiol.
1993
, vol. 
154
 (pg. 
566
-
572
)
40
Zhu
G. H.
Huang
J.
Bi
Y.
Su
Y.
Tang
Y.
He
B. C.
He
Y.
Luo
J.
Wang
Y.
Chen
L.
, et al. 
Activation of RXR and RAR signaling promotes myogenic differentiation of myoblastic C2C12 cells
Differentiation
2009
, vol. 
78
 (pg. 
195
-
204
)
41
Filling
C.
Wu
X.
Shafqat
N.
Hult
M.
Martensson
E.
Shafqat
J.
Oppermann
U.C.T.
Subcellular targeting analysis of SDR-type hydroxysteroid dehydrogenases
Mol. Cell Endocrinol.
2001
, vol. 
171
 (pg. 
99
-
101
)
42
Markus
M.
Husen
B.
Adamski
J.
The subcellular localization of 17β-hydroxysteroid dehydrogenase type 4 and its interaction with actin
J. Steroid Biochem. Molec. Biol.
1995
, vol. 
55
 (pg. 
617
-
621
)
43
Luff
A. R.
Atwood
H. L.
Changes in the sarcoplasmic reticulum and transverse tubular system of fast and slow skeletal muscles of the mouse during postnatal development
J. Cell Biol.
1971
, vol. 
51
 (pg. 
369
-
383
)
44
Eisenberg
B. R.
Kuda
A. M.
Peter
J. B.
Stereological analysis of mammalian skeletal muscle. I. Soleus muscle of the adult guinea pig
J. Cell Biol.
1974
, vol. 
60
 (pg. 
732
-
754
)
45
Franzini-Armstrong
C.
Peachey
L. D.
Striated muscle contractile and control mechanisms
J. Cell Biol.
1981
, vol. 
91
 (pg. 
166
-
186
)
46
Cala
S. E.
Jones
L. R.
Rapid purification of calsequestrin from cardiac and skeletal muscle sarcoplasmic reticulum vesicles by Ca2+-dependent elution from phenyl-sepharose
J. Biol. Chem.
1983
, vol. 
258
 (pg. 
11932
-
11936
)
47
Voet
D.
Voet
J. G.
Biochemistry
2004
New York
John Wiley & Sons
(pg. 
800
-
802
)
48
Cherednichenko
G.
Zima
A. V.
Feng
W.
Schafer
S.
Blatter
L. A.
Pessah
I. N.
NADH oxidase activity of rat cardiac sarcoplasmic reticulum regulates calcium induced calcium-release
Circ. Res.
2004
, vol. 
94
 (pg. 
478
-
486
)
49
Zima
A. V.
Copello
J. A.
Blatter
L. A.
Differential modulation of cardiac and skeletal muscle ryanodine receptors by NADH
FEBS Lett.
2003
, vol. 
547
 (pg. 
32
-
36
)
50
Zhou
H.
Yamaguchi
N.
Xu
L.
Wang
Y.
Sewry
C.
Jungbluth
H.
Zorzato
F.
Bertini
E.
Muntoni
F.
Meissner
G.
Treves
S.
Chracterization of recessive RYR1 mutations in core myopathies
Hum. Mol. Genet.
2006
, vol. 
15
 (pg. 
2791
-
2803
)
51
Zhou
H.
Jungbluth
H.
Sewry
C. A.
Feng
L.
Bertini
E.
Bushby
K.
Straub
V.
Roper
H.
Rose
M. R.
Brockington
M.
, et al. 
Molecular mechanisms and phenotypic variation in RYR1 related congenital myopathies
Brain
2007
, vol. 
130
 (pg. 
2024
-
2036
)
52
Ghassemi
F.
Vukcevic
M.
Xu
L.
Zhou
H.
Meissner
G.
Muntoni
F.
Jungbluth
H.
Zorzato
F.
Treves
S.
A recessive ryanodine receptor 1 mutation in a CCD patient increases channel activity
Cell Calcium
2009
, vol. 
45
 (pg. 
192
-
197
)
53
Rios
E.
The cell boundary theorem: a simple law of the control of cytosolic calcium concentration
J. Physiol. Sci.
2010
, vol. 
60
 (pg. 
81
-
84
)
54
Stefanik
P.
Macejova
D.
Mravec
B.
Brtko
J.
Krizanova
O.
Distinct modulation of a gene expression of the type 1 and 2 IP3 receptors by retinoic acid in brain areas
Neurochem. Int.
2005
, vol. 
46
 (pg. 
559
-
564
)
55
Deelman
L. E.
Jonk
L.J.C.
Henning
R. H.
The isolation and characterization of the promoter of the human type I inositol 1,4,5-trisphosphate receptor
Gene
1998
, vol. 
207
 (pg. 
219
-
225
)
56
Haugen
B.
Jensen
D. R.
Sharma
V.
Pulawa
L. K.
Hays
W. R.
Krezel
W.
Chambon
P.
Eckel
R. H.
Retinoid X receptor γ-deficient mice have increased skeletal muscle lipoprotein lipase activity and less weight gain when fed a high-fat diet
Endocrinology
2004
, vol. 
145
 (pg. 
3679
-
3685
)
57
Sleeman
M. W.
Zhou
H.
Rogers
S.
Ng
K. W.
Best
J. D.
Retinoic acid stimulates glucose transporter expression in L6 muscle cells
Mol. Cell. Endocrinol.
1995
, vol. 
27
 (pg. 
161
-
167
)

Supplementary data